To study the mechanism of the IL-4 induced switch from synthesis of immunoglobulin mu to epsilon in human lymphocytes, we have continued our analysis of PCR-amplified composite mu-epsilon switch regions formed by the switch recombination. From the sequence of the composite switch region in the cell line 2Cl0, we identified an insertion at the mu-epsilon junction that apparently derived from the gamma1 gene switch region. This insertion implies that the 2Cl0 line switched initially to gamma1 and subsequently to epsilon. Our finding that similar insertions occur with some frequency in fresh human lymphocytes switching to epsilon in culture led us to ask whether such sequential switching was obligatory and whether it always involved gamma1. To approach these questions, we have developed PCR assays for the reciprocal products of the isotype switch recombination. Preliminary evidence suggests that (1) the sequential switch can involve other gamma genes besides gamma1, and (2) sequential switching is not obligatory, as some reciprocal products indicative of a direct mu-to-epsilon switch are observed. Our collaborative study of epsilon mRNA transcripts in cells switching to epsilon has led to the identification of two alternative mRNA splice forms. One of these contains a putative transmembrane (TM) region but is unusual in having a very large extracellular segment between the last immunoglobulin domain and the TM region; this segment could be important for the signal transduction machinery of epsilon-expressing lymphocytes. The second form includes the second of two "membrane" exons but this is read in an alternative reading frame; this form lacks a hydrophobic segment and would presumably be translated into a secreted protein that would be 134 amino acids larger than the classical secreted form. We have obtained Western blot evidence for an epsilon-related protein of the size expected for this larger secreted form and are currently attempting to generate antibodies specific for the C-terminal end predicted for this larger form in order to verify the existence of the protein and study its function.